Don’t miss out on the latest CyclingTips updates.
It’s been a couple months now since dockless bikeshare company oBike started pulling out of Melbourne. That withdrawal came barely a year after oBike’s first started operating in the city — a year in which the Singapore-based company seemed to receive near constant criticism from local government, the media, and citizens alike.
But now that the dust has settled on oBike’s Melbourne experiment, it’s becoming clearer what the whole scheme was actually about. oBike might have been under fire for cluttering our streets with bikes that seemingly no one wanted to ride, but as Peter Chambers from RMIT University’s School of Global, Urban and Social Studies writes, the real story of oBike has little to do with bikes or cycling at all.
When oBike left Melbourne in June 2018, the city’s mayor, Sally Capp, alleged that the scheme’s failure was due to “the behaviour of people using the oBikes”, a variant of “this is why we can’t have nice things”, which popped up elsewhere. But were these things nice? Did “we” ever “have” them?
oBike’s year of failure in Melbourne has lead to widespread dumping — the company’s real legacy for the city — and a serious issue that hasn’t been addressed. But in order to understand this, we first have to get at the reasons why oBike failed.
It’s notable that a scheme like oBike only works once smartphone uptake is almost ubiquitous — as in Australia — and where most people who have smartphones also have credit cards whose details they’re willing to share online for ease of payment. That’s a harder sell in other places, such as Germany and Japan.
There are of course parallels here with Uber, but suffice to say that, from an end-user point of view, as a “disruption” of car-based ride services, when Uber launched it was a better product design and service than the taxis that preceded it. For users in Melbourne, and especially in Sydney, where getting anywhere is expensive and time-consuming, Uber was an undeniable improvement on stubbornly dysfunctional transport arrangements – though hardly the solution to gridlock.
Could oBike in any meaningful way be said to offer a similar disruption for bikes? If we’re thinking about the “future of transportation” claim, it bears acknowledging that neither oBike’s product design, nor the bikes themselves, help to solve any existing structural transport problems in Sydney and Melbourne, for a number of reasons. Rather, like other dockless bike hire schemes — often mistakenly called “sharing” — it works on the more modest idea of the “final mile”.
The “final mile” relies on transport that can’t quite get you to your destination. In Australia’s middle and outer suburbs, and perhaps Canberra and Adelaide, it makes a kind of sense. But in suburban areas where commuters might benefit from a service that saves them a 30-minute “final mile” walk home at the end of a long day, is there the density to make it viable? Do you have a system design and staff capability to deal with the dispersion of your fleet? And are the units designed to withstand mischief at stations overnight?
In the inner cities where oBike chose to launch in Australia, the very existence of a “final mile” problem was highly questionable. Sydney is hilly and has narrow footpaths and streets with fiercely bad traffic. Melbourne’s CBD has a free tram zone. Nice or no, this, perhaps, was one reason why “we” didn’t need these things.
But what of the degree of niceness of the bikes, and the system that enables them to, you know, work nicely?
oBike relied on an urban operating environment where large numbers of customers have a smartphone – with Bluetooth and a camera that can read QR codes – and are willing and able to download an app, enter their credit card details, and, in oBike’s case, pay the $69.99 security deposit to access the service. These are the preconditions of using the platform. But could you even use oBike’s platform if you wanted to? And what happened if you tried, via the app?
A scroll through the feedback in the app store showed that getting started proved difficult for many. If the QR codes didn’t scan the bike wouldn’t unlock. Sometimes the lock didn’t work, but the system had begun charging the customer. Sometimes the bike wouldn’t re-lock, so charges continued. And very often the bike wasn’t where the app indicated it could be found, so you had no transport service.
This is really important, as the whole idea of “final mile” relies on bikes being there, ready, and able to be spontaneously used without too much faffing around. If you can’t say it “beats walking” then it doesn’t solve any problem at all; it just creates new ones.
App store feedback also showed that customer service was insufficient (to late 2017) or absent (from early 2018 on), and some people reported that getting their deposit back was so difficult you’d think it was designed that way. But even if, say, you’d left the pub and it suddenly began to rain, or public transport was cancelled and you were cycling distance from home and saw an oBike and would love to use it and then tried to, the system, as designed, defied you to use it as intended – especially for the following reason.
The issue of locations gets us to the nub of a very basic problem with oBike’s product design. As the bikes had no internal GPS (unlike Ofo, which has now also withdrawn from Australia), the company had no way of knowing where its fleet was drifting through the city over time, if people didn’t comply with its way of doing things. This meant that oBike’s “nice things” quickly became invisible to the very entity responsible for keeping them nice. This is why we can’t have nice things – because we didn’t know where they were.
By early 2017 I became concerned that large numbers of oBikes were being dumped in my local creek, indicating a systemic ecological issue for which no one was taking responsibility. During one sweep along the Moonee Creek on March 27, between Footscray Rd and Strathmore High — a roughly 10km stretch — I counted 25 units in and around the waterway. Of those I counted, 44% were in the creek, a further 28% showed signs of having been fished out, and 0% were on the system, roadworthy, and fitted with a helmet.
In part, this was because the company had appeared to leave them in local parks abutting the creek — a questionable tactical choice of location. But daily commuting also showed me that, in Flemington and Kensington — which are hilly — month by month, the 2017-placed bikes made their way down the hills, and into Melbourne’s waterways.
The upshot of company policy, gravity, and mischief meant that by early 2018, even if you could find an oBike in the place where the system said it would be – which was vanishingly rare – it was also unlikely that it’d resemble a nice thing, let alone be a legally rideable bicycle. Not nice things, not useful things, just dumped things, a big waste and a big mess.
To give one snapshot of this rusting rot and its surreal spectres, all one need do is open the app – still running in the weeks after oBike had withdrawn from operations. In those parts of my neighbourhood of Flemington and Kensington, the app shows no less than six oBikes available. None of these bikes exist where the company says they still are; the only one that actually is there wasn’t on the system when I checked.
In back streets and parks near my house I can take you to a few I pass daily that, as far as I know, never moved since they were left there in June 2017, though now they are useless, rusting, and vandalised. Again, only one of these ever had a helmet.
Yet the app continued to be updated after operations ceased, and asked users to update to keep using the platform. The June-July 2018 update won’t get you a bike, but apparently it does prevent customers from getting their security deposit back.
If the system doesn’t take account of local conditions, isn’t useful as transport for those able and keen to try it, and if the product design is systemically faulty for end users and the company, what’s the business model?
This is where we come to the essence of oBike, missed by nearly everyone across its year in Melbourne, from the upbeat reporting last June to Melbourne’s mayor this June.
oBike is a vehicle, yes, but of a different order. It’s a platform, true, but not for riding – for venture capital. This is well captured by Sun Sheng Han in one of the only enquiries that considered how the scheme might actually work. It’s worth clicking through and having a read of his great analysis, then coming back to me.
Let’s start with the nice things of this business model. The manager of another dockless bike hire program in Australia gave me an estimate of the unit cost as somewhere around the $50 mark. If you spend some time searching on Alibaba, and consider the size of an order for global operations needing tens of thousands of units — or more — it could be lower.
The initial fleet in Melbourne was 1,200 bikes, rising to 2,500. The security deposit was $69.99 – above probable unit cost. If this is true, it means the deposit amply covered the cost of production, hedging against damage or loss, because one new user pays for a unit. This immediately rendered marginal any caring for or collecting damaged, vandalised, lost, or impounded bikes: services that a bike scheme that was about bikes would do, but that one that was about venture capital never needed to bother with.
These were never nice things, they were always cheap things, merely as good as they needed to be to snag the security deposit a dozen or so times, then be replaced.
These effects of oBike’s business decisions speak to the ineffectual nature of the memorandum of understanding (MOU) drawn up between the City of Melbourne and Port Phillip in August and October 2017, where a $50 collection fee was to be levied by councils for each impounded bike. But low unit costs and wrist-slap fines meant that paying contractors to pick up impounded units, even large numbers of them, was a big “why bother?”
It also explains why it wouldn’t make good business sense to retain concerned, skilled staff to place, re-place, and repair the fleet, or answer customer queries (as Ofo did seek to do in Adelaide). From the get-go, the way oBike designed their scheme removed any incentive for its employee (yes, you read that right – we’ll get back to this) to do anything to prevent the fleet from being misplaced, vandalised, dumped and abandoned. The response to this was weak and then untimely regulation, which provided no meaningful consequence while it could still matter.
How the hell does this make sense, you ask? Of course, you get a lot of data, and you can sell that. Yet even this was bungled. In Europe, oBike’s European customer data was shown to be visible without any protection or encryption. But as Sun Sheng Han enjoins us, let’s focus on the user base as a sum of capital which earns interest, can be invested, and can attract further venture capital, as oBike’s call for business partners shows.
oBike secured an injection of capital – US$30 million – on August 30, 2017. Han writes that “Mobike [ed. another dockless bikeshare scheme] reportedly had over 100 million registered users in June” of last year, generating a $6 billion pool of capital. Assuming oBike were gunning for a deposit from a user base one-tenth the population of Sydney and Melbourne – that’s still just under one million users. If each of them paid $69.99, you just amassed nearly $70 million in venture capital, from an operation where your fleet cost perhaps $120,000, at most.
But hey, if you got to 14,000 users, you just generated $1 million, and if your wage bill covers one core employee … here’s the kicker.
The surreal truth is that oBike apparently had just one full-time employee for Australian operations – Lim Chee Ping. If you go back and read Doyle’s comments from around the time of the MOU, the City of Melbourne seemed aware of this. At peak controversy, during the Birrarung oBike fishing episode in late September, the company appeared to be headed in Melbourne by its spokesman Chethan Rangaswamy. But Rangaswamy ghosted oBike sometime late in the year – his LinkedIn profile shows he is now the head of marketing for a Blockchain Initiative in Silicon Valley.
It’s impossible to know this for sure, but it seems that, by early 2018, oBike did not have a meaningful human presence in Melbourne, a city it was still operating in. As with the many bikes visible on its system — but nowhere to be found when you wanted to ride one — or in the literally dozens in my local creek — which nobody seems to take responsibility for — the real story of oBike is a story of ghosts in a city filled with large yellow bikes that no one can ride. Neither “nice things”, or things that “we have”, except as hard rubbish we’re left to clean up.
The dumping we’re left with in Melbourne brings me to my final point, one that’s close to my heart as a cyclist and a resident. When oBike left, the widespread view was that dumping happened here because of “the behaviour of users”. That is, “we” “couldn’t have” “nice things” because of something uniquely naughty about Australians, a strange cultural cringe explanation that holds Melbourne and Sydney’s residents accountable for acts of mischief that the overwhelming majority had zero involvement in.
Australian reader: did you ever touch one? Crucially, dumping was reported in Singapore months before it happened in Melbourne — it was forseeable and should have been planned for. Dumping continues to be an issue facing all dockless bike hire schemes in mainland China, and dumping, clutter and nuisance concerns have plagued dockless scooter hire schemes in California.
So far, dockless has been characterised by a stubborn absence of demand relative to deafly enthusiastic supply. Alongside this mismatch, we have dumping as the common effect of dockless: a systemic and foreseeable issue generated by all schemes to date, true for any city where we live with strangers and lock up our nice things (see the cable and U-lock on my Colnago Master, as one example precious to this author.)
Given all the factors I’ve described, it’s time to acknowledge that disgraced mayor Doyle was right all along, and knew something — oBike was never about the bike, because it was “not a bike-share scheme”, it was “a way to generate enormous amounts of venture capital … a kind of a shadowy organisation … privatising the public realm”, not paying tax, without employees.
The evidence on just how shadowy is beginning to trickle in. Just this month, the Straits Times reported that oBike Singapore abruptly transferred 10 million Singapore dollars to Hong Kong, while liquidation proceedings were active.
For someone like me, who loves bikes, their city and its good people, it’s disappointing to hear people still blaming my city and its people for oBike’s failure, without engaging with what we know and can surmise, and without considering the strange mess that the people who actually live here are now left to deal with, but haven’t.
As one of the more than perhaps four million Melburnians who never laid a foot or a hand on an oBike, I hope the story of the company’s Melbourne venture is something we can learn from.
About the author
Pete Chambers is a senior lecturer in criminology and justice in the School of Global Urban and Social Studies at RMIT. He is currently writing a book exploring the deep roots of conflict between motorists and cyclists in global cities. He commutes every day by bike and is an average road cyclist who is better at riding into headwinds than up hills. He favours lugged frames and fine ales.